Free Republic
Browse · Search
Bloggers & Personal
Topics · Post Article

To: CheshireTheCat
Jim Hoft doesn't understand p values and statistical significance. P values less than 0.05 are statistically significant.

Here is the report: Association of State-Issued Mask Mandates and Allowing On-Premises Restaurant Dining with County-Level COVID-19 Case and Death Growth Rates — United States, March 1–December 31, 2020


18 posted on 03/07/2021 8:04:31 AM PST by Arones (People say "this is not who we are." Have you not turned on the tv since Memorial Day? -Mark Steyn)
[ Post Reply | Private Reply | To 1 | View Replies ]


To: All

What is the point, here?

The virus has killed 21% more old people in the past 12 months than typically die in a pre-virus year. 21%!!

How high must that number go before we would approve of measures to reduce it? Does the mask measure not work? Maybe not. Then what measure will you accept? Any at all?

Or you can just announce you are okay with 21% increases in elderly death. Good luck getting elected on that platform. Old people vote. The ones you left alive, I mean.


21 posted on 03/07/2021 8:25:42 AM PST by Owen
[ Post Reply | Private Reply | To 18 | View Replies ]

To: Arones

Thank you for posting this original study. I trust NOTHING from Gateway Pundit.

This is yet another great example of why.

They take one random incorrect tweet and turn it into a story that gets broadcasts by conservatives, who are then embarrassed by smug liberals who point out the obvious flaw.

I’m starting to believe that GP is actually a liberal site, designed for this purpose.


28 posted on 03/07/2021 8:44:24 AM PST by SomeCallMeTim ( The best minds are not in government. If any were, business would hire them!it)
[ Post Reply | Private Reply | To 18 | View Replies ]

To: Arones
Not sure what Jim Hoft understands.

As I read the study it seems like hogwash though. The p-value is based on measure the change of the change of the number of cases around the time the mask mandates were given.

Now of course, p-values are a measure of how different data sets vary form each other, with a low value indicating there is more significant a difference. But the difference should be between a known expect result (null hypothesis) and the result that is true if the hypothesis tested is true: in this case that the masks are helping reduce the change of the change in cases and deaths.

The problem I have with this study is that it does not seem competent to expect a null hypothesis of no relation between mask mandates and changes to the changes in cases and deaths, even if the masks did not help at all. There were wild differences in the number of cases in different areas at different times with places where there were no mask mandates being far lower in the number of cases. That places where the virus was just starting out might have a positive second derivitive of cases/deaths with respect to time opposed to those where the virus was already infecting people is common sense. It is what I would expect.

Does not the chart of cases over time not usually look like a wave with it accelerating upwards then slowing down, flattening at the top and then declining? Why would one expect that there was no relation between where one is on this chart and what policies might be put in place?

30 posted on 03/07/2021 8:52:25 AM PST by AndyTheBear
[ Post Reply | Private Reply | To 18 | View Replies ]

To: Arones
The p values are not relevant — this study is a large can of garbage.

The continued absence of any Gold Standard research (randomized double blind controlled experiments) demonstrating the efficacy of masks is rather telling.

Two critical variables not controlled in this study are the virulence of the virus and the susceptibility of the individuals within the populations.

34 posted on 03/07/2021 9:19:03 AM PST by PsyCon
[ Post Reply | Private Reply | To 18 | View Replies ]

To: Arones; AndyTheBear; CheshireTheCat
Jim Hoft doesn't understand p values and statistical significance. P values less than 0.05 are statistically significant.

Here is the report: Association of State-Issued Mask Mandates and Allowing On-Premises Restaurant Dining with County-Level COVID-19 Case and Death Growth Rates — United States, March 1–December 31, 2020

Thanks for the detective research that was neglected. However, notice the specious comparison in the CDC graphic:

Requiring masks and allowing indoor dining are two different scenarios. Indoor dining means masks are not worn in close proximity regardless of typical mask mandates, and is simply not the same as not requiring masks in general. Here in MA the governor requires masks to be worn at all times when outside on public property, regardless of the spacing, and which is absurd and no stats support that unequivocal mandate (and a Oct 29, 2020 Massachusetts report on clusters found that household transmission (via tracing) was at the center of the vast majority (almost 94 percent) of recent COVID-19 cases). Yet this forced fallacious comparison is what the CDC uses to advocate for universal mask mandates:

Allowing any on-premises dining at restaurants was associated with increases in daily COVID-19 case growth rates 41–60, 61–80, and 81–100 days after reopening, and increases in daily COVID-19 death growth rates 61–80 and 81–100 days after reopening. Implementing mask mandates was associated with reduced SARS-CoV-2 transmission, whereas reopening restaurants for on-premises dining was associated with increased transmission.

Rather, this study does not substantiate that implementing mask mandates would reduce rates (although I am sure they do somewhat, and the more the better until you cannot breath at all) since they take them off while eating, and thus this data simply does not support general mask mandates unless indoor dining and the like are disallowed. And therefore the sophistry that the CDC is engaging in here impugns its integrity in general, and results in typical misleading parroting MSM headlines (which on mobile devices is all that many sheep read) such as,

CDC study shows link between mask mandates, reduced COVID ... https://abcnews.go.com/Health/cdc-study-shows-link 1 day ago

Coronavirus mask mandates linked to decline in cases ... https://www.foxnews.com/health/coronavirus-mask. 1 day ago

Mask mandates tied to fewer COVID-19 cases, deaths, U.S ... https://www.msn.com/en-xl/news/other/mask-mandates

CDC study finds mask mandates reduced COVID-19 cases and ... https://theweek.com/speedreads/970549/cdc-study... 1 day ago

Mask Mandates Linked To Decrease In COVID-19 Cases, Study ... https://www.npr.org/sections/coronavirus-live...

CDC study shows link between mask mandates, reduced COVID ... https://www.goodmorningamerica.com/news/story/cdc... 1 day ago

This is akin to the parroting misleading headlines asserting the absurdity,

"Researchers find face masks don't hinder breathing during exercise" (FR thread):

39 posted on 03/07/2021 10:02:54 AM PST by daniel1212 (Turn to the Lord Jesus as a damned+destitute sinner, trust Him to save + be baptized + follow Him!)
[ Post Reply | Private Reply | To 18 | View Replies ]

To: Arones

The GP article quotes from a tweet.

“The results were inside the statistical margin of error:

The results are inside the margin for statistical error

— Breaking911 (@Breaking911) March 6, 2021”


50 posted on 03/07/2021 2:22:39 PM PST by ifinnegan ( Democrats kill babies and harvest their organs to sell)
[ Post Reply | Private Reply | To 18 | View Replies ]

Free Republic
Browse · Search
Bloggers & Personal
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson